I will preface this by saying that I am a Minecraft modpack developer; I have no idea if the following issue applies to mods, but I'm certain it applies to Minecraft modpacks.
As we all know, the Curseforge website offers end-users five sorting options for discovering new modpacks: Date Created, Last Updated, Name, Popularity, and Total Downloads. For the context of end-users discovering modpacks, these options can generally be divided into two categories; Date Created and Last Updated are useful for finding content that's brand new but with absolutely no guarantee of quality. By critically stark contrast, Popularity and Total Downloads bring you modpacks that are already successful, ones you've heard of. Unfortunately, this sharp divide makes it difficult and usually even impossible for new content to make it past a couple thousand downloads without an external source of exposure. See, the current model seems to expect new content to pick up no less than tens of thousands of downloads before being featured in the Popularity and Total Downloads tabs, yet there aren't tens of thousands of downloads to be earned in the Date Created and Last Updated tabs. This forms an impassable divide between existing modpacks with existing userbases and new modpacks with potentially revolutionary content. This divide isn't good for anyone; it's unfair to pack developers and, more importantly, it keeps fresh ideas buried and unreachable.
So, what is to be done? I believe a number of things could help greatly:
A new "Rising" sorting option that sorts content by downloads per day but has an exponential dropoff for increasing total downloads. This would expose promising modpacks and favor those that make good use of the exposure they do have.
The two other potential solutions I see, a simple rating system or Reddit-style vote system, have been suggested in other idea tickets and need no explanation here.